Development Of Intelligent Fire Alarm System Based On Embedded Platform

The integrate of fire alarm system and embedded technology has become an inevitable trend, as people increasingly high demand for fire alarm system and the development of embedded technology.Through the embedded software programming,not only the on-site fire information could be monitored by people in the more intuitive and more accurate way,but also the performance of operability, stability, real-time of the monitoring system could be improved.At the same time.through the embedded Web technology, not only the on-site fire information could be accessed in a real-time way, but also the effective protection for centralized monitoring of fire information could be provided.The intelligent fire alarm system,as the object of study.would be discussed in this paper.and the embedded platform with an piece of ARM9microprocessor and embedded Linux operating would be constructed. Through the embedded platform,the temperature could be monitored regionally by Qt embedded software technology.and could be monitored together by the embedded Web technology.In addition.the algorithm for the judgment of fire information would be studied in the paper.The main research work of the paper consists of the following several parts specifically.1.The paper discusses the neural network ensemble Adaboost.RT algorithm and SVR algorithm for determining the fire information.The Adaboost.RT algorithm is studied,several BP neural network is ensembled, the simulation comparison and analysis are made between a single BP neural network and Adaboost.RT.SVM(support vector machine) is studied, SVR algorithm is used in fire prediction.and simulation and analysis are also made through data.2.The DS18B20driver programming is compiled.The temperature acquisition principle of DS18B20temperature sensor and control method are analyzed.The programming methods of the character device driver are learned.and then the DS18B20driver is written and compiled into the Linux2.6.12kernel,providing an interface for upper layer software to read the scene temperature.3.The Qt2embedded temperature monitoring software is compiled.The Qt2program build environment is built.the temperature monitoring program is written.The SQLite embedded database is transplanted to the embedded Linux platform.4.The Web network remote monitoring system is built.A Boa embedded Web server is transplanted.The websites of the user login and remote temperature acquisition are made.CGI programs that respond to the remote access to the operation of the SQLite database are written.5.The intelligent fire alarm system is tested and analyzed.The DS18B20driver,the Qt2temperature monitoring software,the CGI program are tested and analyzed respectively,for the overall realization of intelligent fire alarm system function.